ietf-smime
[Top] [All Lists]

RE: Jim's Comments: draft-ietf-smime-cmsalg-00

2001-07-10 20:04:23

"Jim Schaad" <jimsch5(_at_)home(_dot_)com> writes:

If this is the preferred encoding, it is certianly not displayed as such in
the current text.  If you really think that is true then we need to make the
following change:

"Implementations SHOULD generate SHA-1 AlgorithmIdentifiers as absent".

Wouldn't it be better to just include the historical note about
AlgorithmIdentifier parameters and let implementors decide on what to do?
Telling people why it's the way it is would help reduce a lot of confusion when
implementors who aren't aware of this piece of ASN.1 trivia find both forms in
use, in my drafts I've included the correct (or more precisely currently
fashionable) form, but with a note to say that people should be prepared to
encounter the other alternative (I'm happy with SHOULD, I'd just like to see a
caveat included so that every new implementor who comes along doesn't fall into
the same trap).

Peter.